Testing experimental designs in liquid chromatography (I): Development and validation of a method for the comprehensive inspection of experimental designs

Enhancing LC×LC separations through multi-task Bayesian optimization

Method development in comprehensive two-dimensional liquid chromatography (LC×LC) is a challenging process. The interdependencies between the two dimensions and the possibility of incorporating complex gradient profiles, such as multi-segmented gradients or shifting gradients, make trial-and-error method development time-consuming and highly dependent on user experience. Retention modeling and Bayesian optimization (BO) have been proposed as solutions to mitigate these issues. However, both approaches have their strengths and weaknesses. On the one hand, retention modeling, which approximates true retention behavior, depends on effective peak tracking and accurate retention time and width predictions, which are increasingly challenging for complex samples and advanced gradient assemblies. On the other hand, Bayesian optimization may require many experiments when dealing with many adjustable parameters, as in LC×LC. Therefore, in this work, we investigate the use of multi-task Bayesian optimization (MTBO), a method that can combine information from both retention modeling and experimental measurements. The algorithm was first tested and compared with BO using a synthetic retention modeling test case, where it was shown that MTBO finds better optima with fewer method-development iterations than conventional BO. Next, the algorithm was tested on the optimization of a method for a pesticide sample and we found that the algorithm was able to improve upon the initial scanning experiments. Multi-task Bayesian optimization is a promising technique in situations where modeling retention is challenging, and the high number of adjustable parameters and/or limited optimization budget makes traditional Bayesian optimization impractical.

A perspective on the use of deep deterministic policy gradient reinforcement learning for retention time modeling in reversed-phase liquid chromatography

Artificial intelligence and machine learning techniques are increasingly used for different tasks related to method development in liquid chromatography. In this study, the possibilities of a reinforcement learning algorithm, more specifically a deep deterministic policy gradient algorithm, are evaluated for the selection of scouting runs for retention time modeling. As a theoretical exercise, it is investigated whether such an algorithm can be trained to select scouting runs for any compound of interest allowing to retrieve its correct retention parameters for the three-parameter Neue-Kuss retention model. It is observed that three scouting runs are generally sufficient to retrieve the retention parameters with an accuracy (mean relative percentage error MRPE) of 1 % or less. When given the opportunity to select additional scouting runs, this does not lead to a significantly improved accuracy. It is also observed that the agent tends to give preference to isocratic scouting runs for retention time modeling, and is only motivated towards selecting gradient scouting runs when penalized (strongly) for large analysis/gradient times. This seems to reinforce the general power and usefulness of isocratic scouting runs for retention time modeling. Finally, the best results (lowest MRPE) are obtained when the agent manages to retrieve retention time data for % ACN at elution of the compound under consideration that spread the entire relevant range of ACN (5 % ACN to 95 % ACN) as well as possible, i.e., resulting in retention data at a low, intermediate and high % ACN. Based on the obtained results, we believe reinforcement learning holds great potential to automate and rationalize method development in liquid chromatography in the future.

Experimental design and re-parameterization of the Neue-Kuss model for accurate and precise prediction of isocratic retention factors from gradient measurements in reversed phase liquid chromatography

The present work describes a re-parameterization of the Neue Kuss (NK) model for describing retention in liquid chromatography, and this re-parameterized model is used to fit a large set of isocratic retention measurements with improved convergence properties relative to the original parameterization of the model. Next, an experimental design for retention measurements using mobile phase gradient elution conditions is proposed for the purpose of obtaining accurate and precise NK parameters. Simulated retention data for mobile phase gradient elution conditions with two different levels of noise, as well as an essentially zero noise level were fit with the re-parameterized model. The results showed that the re-parameterized fits yielded average (absolute value) prediction errors for the parameters at the highest noise level of 7.2 % for S1,ref, 18 % for S2,ref and 6.2 % for kref (the re-parameterized NK model parameters). These errors were significantly smaller than those for the original parameterization of the NK model, where the errors were 23 % for S1, 25 % for S2 and 160 % for kw (the original NK model parameters). Furthermore, isocratic retention factors predicted using these model parameters were found to have an average magnitude of error of 0.51 % for the re-parameterized model, as opposed to 6800 % for the model with the original parameterization. A further test of this approach was carried out for independent experimental measurements for five solutes on a C18 column. The average magnitude of error of the isocratic retention factors predicted from parameters obtained from fits of gradient data was 1.6 %, provided that the range of organic solvent compositions that the solute sampled in the mobile phase gradient experiments was consistent with the isocratic experiments. These results indicate that the re-parameterization of the NK model allows for significant improvements in the fitting process, and that the proposed experimental design allows for NK parameters to be extracted from mobile phase gradient experiments, with prediction accuracies of isocratic retention factors on the order of 1-2 %.

Performance of global retention models in the optimisation of the chromatographic separation (I): Simple multi-analyte samples

Conventional retention models lead to accurate descriptions of the elution behaviour from the fitting of data for single solutes or from a set of solutes, one by one. However, the simultaneous fitting of several solutes through a regression process that separates the contributions of column and solvent from those of each solute is also possible. The result is a global retention model constituted by a set of equations with some common parameters (those associated with column and solvent), whereas others, specific to each solute, differ for each equation. This work explores the possibilities, advantages, and limitations of global models when they are applied to the optimisation of chromatographic resolution. A set constituted by 13 drugs (diuretics and β-blockers) and a training experimental design of seven multi-linear gradients are considered. Since standards for all compounds were available, the optimisation based on global models could be compared with the conventional optimisation, which is based on individual models. In their current state, global models do not predict changes in elution order, but they do allow for incorporating additional solutes (e.g., new analytes or matrix peaks) with only one new experiment. This possibility is explored by extending the model for the 13 analytes to include 26 peaks associated with a contamination in the injector. The combination of individual and global models allows an optimisation where the effects of matrix peaks on the separation of analytes can be integrated.

Accuracy of retention model parameters obtained from retention data in liquid chromatography

In liquid chromatography (LC), it is often very useful to have an accurate model of the retention factor, k, over a wide range of isocratic elution conditions. In principle, the parameters of a retention model can be obtained by fitting either isocratic or gradient retention factor data. However, in spite of many of our own attempts to accurately predict isocratic k values using retention models trained with gradient retention data, this has not worked in our hands. In the present study we have used synthetic isocratic and gradient retention data for small molecules under reversed-phase LC conditions. This allows us to discover challenges associated with predicting isocratic k's without the confounding influences of experimental issues that are difficult to model or eliminate. The results indicate that it is not currently possible to consistently predict isocratic retention factors for small molecules with accuracies better than 10%, even when using synthetic gradient retention data. Two distinct challenges in fitting gradient retention data were identified: 1) a lack of 'uniqueness' in the parameters; and 2) an inability to find the global optimum fit in a complex fitting landscape. Working with experimental data where measurement noise is unavoidable will only make the accuracy worse. This article is protected by copyright. All rights reserved.

Testing experimental designs in liquid chromatography (II): Influence of the design geometry on the prediction performance of retention models

In liquid chromatography, the reliability of predictions carried out with retention models depends critically on the quality of the training experimental design. The search of the best design is more complex when gradient runs are used instead of isocratic experiments. In Part I of this work (JCA 1624 (2020) 461180), a general methodology based on the error propagation theory was developed and validated for assessing the quality of training designs involving gradients. The treatment relates the mathematical properties of a retention model with the geometry of the training designs and their subsequent predictions. In that work, only five usual designs were considered. Part II investigates in detail the effects on predictions when the features of the training design (number and distribution of the experiments, initial and final modifier content, gradient slope(s), and location of gradient nodes and pulses) are varied systematically. Several groups of related designs containing one or more isocratic steps, linear or multi-linear gradients, or mixed isocratic/gradient runs, among others (in total 38 designs) were evaluated. Box and whiskers and triple plots of expected relative uncertainties were used to evidence the differences in prediction performance. The purpose was to give recommendations to construct designs with good prediction performance. The best designs sample (considering all runs) concentrations as diverse as possible, at any gradient time.

Closed form approximations to predict retention times and peak widths in gradient elution under conditions of sample volume overload and sample solvent mismatch

Closed form expressions for the prediction of retention times and peak widths for gradient liquid chromatography are particularly useful in understanding, rationalizing and optimizing separations. These expressions are obtained by integrating differential equations, in conjunction with a model of the variation of the retention factor as a function of mobile phase composition. Two of these models, the linear solvent strength (LSS) model and the Neue-Kuss (NK) model are explored in the present work. Here, we expand on these closed form expressions to account for effects of sample volume overload and a mismatch between the sample solvent and the initial mobile phase composition for the gradient. We show that there have been errors in expressions reported in the literature, and we have evaluated the accuracy of the predictions from the closed form expressions reported here using a recently developed liquid chromatography simulator. The expressions assume a constant plate height and consider elution across four zones of the gradient profile - elution in the sample solvent, elution in the initial (isocratic) mobile phase caused by the gradient delay volume, elution during a linear gradient, and elution post-gradient at the final (isocratic) mobile phase composition. The expressions generally give reasonably accurate predictions for retention times and peak widths, except for cases where the solute elutes during transitions between the different zones. The average magnitude of the prediction errors for retention time and peak width relative to simulation were 0.093% and 0.40% for the LSS expressions for ten amphetamine solutes at 36 different separation conditions, and 0.22% and 1.8% for the NK expressions for eight alkylbenzene solutes at 36 different separation conditions, respectively.

Comparison of the Fitting Performance of Retention Models and Elution Strength Behaviour in Hydrophilic-Interaction and Reversed-Phase Liquid Chromatography

Hydrophilic interaction liquid chromatography (HILIC) is able to separate from polar to highly polar solutes, using similar eluents to those in the reversed-phase mode (RPLC) and a polar stationary phase, where water is adsorbed onto its surface. It is widely accepted that multiple modes of interaction take place in the HILIC environment, which can be far more complex than the interactions in an RPLC column. The behaviour in HILIC should be adequately modelled to predict the retention with optimisation purposes and improve the understanding on retention mechanisms, as is the case for RPLC. In this work, the prediction performance of several retention models is studied for seven HILIC columns (underivatised silica, and silica containing diol, amino and sulfobetaine functional groups, together with three columns recently manufactured with neutral, anionic, and cationic character), using uracil and six polar nucleosides (adenosine, cytidine, guanosine, thymidine, uridine, and xanthosine) as probe compounds. The results in HILIC are compared with those that were offered by the elution of several polar sulphonamides and diuretics analysed with two C18 columns (Chromolith Speed ROD and Zorbax Eclipse XDB). It is shown that eight retention models, which only consider partitioning or both partitioning and adsorption, give similar good accuracy in predictions for both HILIC and RPLC columns. However, the study on the elution strength behaviour, at varying mobile phase composition, reveals similarities (or differences) between RPLC and HILIC columns of diverse nature. The particular behaviour for the HILIC and RPLC columns was also revealed when the retention, in both modes, was fitted to a model that describes the change in the elution strength with the modifier concentration.